- Email: [email protected]
Diagnostic and predictive value of urine PCA3 gene expression for the clinical management of patients with altered prostate-specific antigen
Actas Urol Esp. 2014;38(3):150---155
Actas Urológicas Españolas www.elsevier.es/actasuro
ORIGINAL ARTICLE
Diagnostic and predictive value of urine PCA3 gene expression for the clinical management of patients with altered prostate-specific antigen夽 N. Rodón a,∗ , I. Trías a,b , M. Verdú a,b , R. Román a , A. Domínguez c , M. Calvo d , J.M. Banus c , A.M. Ballesta e , M.L. Maestro f , X. Puig a,b,g a
BIOPAT, Biopatologia Molecular, S.L., Grup Assistència, Barcelona, Spain Histopat Laboratoris, S.L., Barcelona, Spain c ICUN, Institut Català d’Urologia i Nefrologia, Barcelona, Spain d Departamento de Estadística, Facultad de Biología, Universidad de Barcelona, Barcelona, Spain e Analiza. S. Análisis Clínicos, Hospital Moncloa, Madrid, Spain f Unidad de Genómica y Reproducción Asistida, Hospital Clínico Universitario San Carlos, Madrid, Spain g Hospital de Barcelona, SCIAS, Grup Assistència, Barcelona, Spain b
Received 25 April 2013; accepted 17 July 2013 Available online 17 December 2013
KEYWORDS PCA3; Prostate-specific antigen; Screening; Prostate cancer; Gleason; Tumor aggressiveness
Abstract Objective: Analyze the impact of the introduction of the study of PCA3 gene in post-prostatic massage urine in the clinical management of patients with altered PSA, evaluating its diagnostic ability and predictive value of tumor aggressiveness. Methods: Observational, prospective, multicenter study of patients with suspected prostate cancer (PC) candidates for biopsy. We present a series of 670 consecutive samples of urine collected post-prostatic massage for three years in which we determined the «PCA3 score» (s-PCA3). Biopsy was only indicated in cases with s-positive PCA3. Results: The s-PCA3 was positive in 43.7% of samples. In the 124 biopsies performed, the incidence of PC or atypical small acinar proliferation was 54%, reaching 68.6% in s-PCA3 ≥ 100. Statistically significant relationship between the s-PCA3 and tumor grade was demonstrated. In cases with s-PCA3 between 35 and 50 only 23% of PC were high grade (Gleason ≥ 7), compared to 76.7% in cases with s-PCA3 over 50. There was a statistically significant correlation between s-PCA3 and cylinders affected. Both relationships were confirmed by applying a log-linear model. Conclusions: The incorporation of PCA3 can avoid the need for biopsies in 54% of patients. s-PCA3 positivity increases the likelihood of a positive biopsy, especially in higher s-PCA3 100 (68.6%). s-PCA3 is also an indicator of tumor aggressiveness and provides essential information in making treatment decisions. © 2013 AEU. Published by Elsevier España, S.L. All rights reserved.
夽 Please cite this article as: Rodón N, Trías I, Verdú M, Román R, Domínguez A, Calvo M, et al. Valor diagnóstico y predictivo del estudio en orina del gen PCA3 para el manejo clínico de pacientes con PSA alterado. Actas Urol Esp. 2013;38:150---155. ∗ Corresponding author. E-mail address: [email protected] (N. Rodón).
2173-5786/$ – see front matter © 2013 AEU. Published by Elsevier España, S.L. All rights reserved.
Diagnostic and predictive value of PCA3 gene in patients with altered prostatic specific antigen
PALABRAS CLAVE PCA3; Antígeno prostático específico; Cribado; Cáncer de próstata; Gleason; Agresividad tumoral
151
Valor diagnóstico y predictivo del estudio en orina del gen PCA3 para el manejo clínico de pacientes con PSA alterado Resumen Objetivo: Analizar el impacto de la introducción del estudio del gen PCA3 en orina posmasaje prostático, en el manejo clínico de pacientes con PSA alterado, valorando su capacidad diagnóstica y predictiva de agresividad tumoral. Métodos: Estudio observacional, prospectivo y multicéntrico de pacientes con sospecha de adenocarcinoma de próstata (AP) candidatos a biopsia. Presentamos una serie de 670 muestras consecutivas de orina posmasaje prostático recogidas durante 3 a˜ nos en las que se determinó el «score de PCA3» (s-PCA3). Se indicó biopsia únicamente en los casos con s-PCA3 positivo. Resultados: El s-PCA3 fue positivo en el 43,7% de las muestras. En las 124 biopsias realizadas, la incidencia de AP o proliferación microacinar atípica fue del 54%, llegando al 68,6% en sPCA3 ≥ 100. Demostramos relación estadísticamente significativa entre el s-PCA3 y el grado tumoral. En los casos con s-PCA3 positivo pero inferior a 50, solo el 23% de AP fueron de alto grado (Gleason ≥ 7); en cambio, en los casos con s-PCA3 superior a 50 lo fueron el 76,7%. La relación entre el s-PCA3 y el porcentaje de cilindros afectados en la biopsia fue significativa. Ambas relaciones se confirmaron al aplicar el modelo log-lineal. Conclusiones: La incorporación del PCA3 permite prescindir de la biopsia en un 54% de los pacientes. La positividad del s-PCA3 aumenta la probabilidad de biopsia positiva, especialmente en s-PCA3 superior a 100 (68,6%). El s-PCA3 es también un indicador de agresividad tumoral y aporta información esencial en la toma de decisiones terapéuticas. © 2013 AEU. Publicado por Elsevier España, S.L. Todos los derechos reservados.
Introduction The generalization of the analysis of prostate-specific antigen (PSA) in serum for early detection of prostate adenocarcinoma (PA) has generated a significant increase in the number of prostate biopsies; however, the low specificity of PSA generates a significant volume of indications for biopsy in patients with benign pathologies. About 75% of the biopsies are negative.1 In the group of patients with altered PSA values but not above 10 ng/ml, which is known as the ‘gray area’ of the diagnosis,2 patients with benign prostatic hyperplasia (BPH), prostatitis, and PA are mixed. In the first 2 cases, the biopsy will be negative and it will be repeated if the PSA levels persist. In the third case, the patients are diagnosed with PA but, in many cases, they are questionable clinical significance carcinomas and in elderly patients.3 In this context, radical surgical treatment does not necessarily modify the patient’s life expectancy.4 In autopsy series reviews, the incidental finding of hidden PA in 30% of patients from 50 to 70 and in 70% after 70 is reported.5 These neoplasms produce no symptoms and do not compromise the patient’s life, but they are able to elevate the serum PSA and trigger the clinical protocol of biopsy and surgical treatment. Currently, there is consensus on the problems generated by PA screening with PSA, and the problems it creates are considered more relevant than the benefits in terms of reducing mortality.1,6 The U.S. agency of preventive medicine (U.S. Preventive Services Task Force) recently issued a grade D recommendation which advises against PA screening with PSA,7 and it has raised reflections and discussions in subsequent articles, which reflect the need to rethink about the current protocols for screening, diagnosis, and treatment of PA and deepen in the search for new specific biomarkers predictive of PA.8,9
The PCA3 gene, widely studied in recent years, is probably one of the first biomarkers which is already under clinical application. This gene, located on chromosome 9q21-22, is transcribed only in prostate tissue and is strongly overexpressed in malignant prostate tissue compared to benign or normal adjacent one.10 In contrast to the PSA, overexpression of PCA3 is specific of PA and independent from prostate volume and the presence of BPH.11 Since 2006, a diagnostic kit of clinical application for the detection of mRNA levels of the PCA3 gene in urine samples collected after prostate massage has been marketed, with which a PCA3 score is obtained (s-PCA3).12 A meta-analysis published in 2010 attributes to PCA3 a negative predictive value of 87.8---98%, recognizing some diagnostic accuracy rates acceptable for use in the diagnosis of PA.13 The test, moreover, is easy to implement both in the clinical protocol and in the laboratory. From the first studies carried out, a directly proportional relationship between the s-PCA3 and the percentage of subsequent positive biopsies became clear.11,14 In some of the most recent studies, we also found a relationship between the s-PCA3 and different determinants of tumor aggressiveness such as the Gleason score, the tumor volume, and extracapsular invasion.11,15,16 The European Association of Urology, in its 2010 Guidelines, includes the use of biomarkers complementary to the PSA, among which the PCA3 stands out due to its higher specificity and sensitivity.17 In this paper we analyze the impact of the introduction of PCA3 in the clinical management of patients with altered PSA, with a special focus on reducing the indication for prostate biopsy and its alleged predictive ability of tumor aggressiveness in patients already diagnosed with PA.
152
Materials and methods We performed an observational, prospective, and multicenter study of patients with suspected PA candidates for prostate biopsy. A protocol of indications with informed consent was established and submitted to an Ethics Committee.
Samples and patients 670 urine samples were included in the study after prostate massage corresponding to 598 consecutive patients presenting to the urologist consultation between October 2009 and November 2012. 66 patients had different urine samples studied throughout their clinical follow-up. For the comparative statistical analyses with other clinical parameters (age, serum PSA rate, presence of prostatitis, and prostate volume) we only used the last study of PCA3, excluding previous ones. This resulted in the exclusion of 72 samples. All the patients met some of the following criteria: elevated PSA and prior negative biopsy, altered PSA but lower than 10 ng/ml without prior biopsy, altered PSA and well-known benign prostatic disease (prostatitis, BPH) and high PSA in patients with comorbidity associated with increased risk in biopsy practice. We excluded patients with previous diagnosis of PA. Age, the value of total PSA (ng/ml), and the prostate volume (cc) of the patients included were recorded. In the cases in which the biopsy was performed, the number of cylinders obtained and, when PA was diagnosed, the number of affected cylinders, the percentage of tumor, and Gleason score were recorded.
PCA3 determination The first voided urine after DRE with prostatic massage, consisting in 3 palpations per lobe, was collected, and the s-PCA3 was determined with the ProgensaTM PCA3Assay (GenProbe) kit. By means of quantification of PCA3 and PSA mRNA, an s-PCA3 was calculated according to the formula: (PCA3 mRNA/PSA mRNA) × 1000. s-PCA3 higher than or equal to 35 was considered positive, interpreted as a high probability of detecting PA in a subsequent biopsy.12 Biopsy was indicated only in cases with positive s-PCA3.
Prostate biopsy protocol Prostate biopsies were performed in the operating room under anesthesia with sedation, ultrasound-guided, and transrectally, obtaining a minimum of 5 cylinders per lobe, with additional samples in the case of finding suspicious nodules on the digital rectal examination or the ultrasound. In those patients who had already had prostate biopsies performed previously, these biopsies were performed by saturation, obtaining a minimum of 10 cylinders per lobe.
Histopathological study Two pathologists performed the histopathological study independently on serial sections (5 microns) of paraffinembedded tissue and stained with hematoxylin-eosin
N. Rodón et al. complemented with immunohistochemical study with racemase (AMACR) and basal cell markers (p63 and 34E12 cytokeratin) when it was necessary. The diagnostic entities covered were PA, atypical small acinar proliferation (ASAP), high-grade prostatic intraepithelial neoplasia (HGPIN), chronic prostatitis, other non-neoplastic processes, or normality. In the PA cases, the grade was determined following the Gleason patterns18 and the number and percentage of affected cylinders were recorded. The ASAP category was reserved for microglandular proliferations with morphological and immunohistochemical criteria of PA (loss of basal layer and/or expression of AMACR) but underrepresented in biopsies to support a definitive diagnosis of PA.
Statistical analysis The possible associations between pairs of categorical variables were analyzed using Fisher’s exact test, while for the continuous variables we used the simple linear regression model. Finally, to determine the association between a shortlist of three categorical variables, the best loglinear model was established using the stepwise algorithm and the Akaike information criterion. The algorithm was based on the full model, i.e., with all possible interactions among the 3 variables.19 All the statistical analyses were performed with the R v.2.15.1 package (©2012, RDevelopmentCoreTeam).20
Results The study included 670 urine samples of 598 patients with a mean age of 65.7 years (standard deviation of 8). In the majority of cases (90.6%), the PCA3 test was indicated due to elevated PSA and negative previous biopsy or due to alteration of PSA not exceeding 10 ng/ml (Table 1). The information rate of the s-PCA3 was 98.2% and the average 47.3. The medians and ranges of the numeric variables studied are shown in Table 2. The s-PCA3 showed no statistically Table 1
Indications for the study of PCA3.
Indication
Number of samples (%)
Negative prior biopsy Elevated PSA ≤ 10 ng/ml Risk factors for biopsy Known benign pathology Not specified
337 (50.3) 270 (40.3) 27 (4.0) 20 (3.0) 16 (2.4)
Total
670 (100)
Table 2
Description of the variables studied. Median (minimum-maximum)
Total PSA (ng/ml) Prostate volume (cc) Number of cylinders studied Percentage of affected cylinders (%)
6 (0.6---134) 53 (10---196) 11 (4---33) 20 (4---100)
Diagnostic and predictive value of PCA3 gene in patients with altered prostatic specific antigen 100
100
90
90
80
80
70 66.6
60 50 40
76.7
77
60 % of cases
70 % of cases
153
33.4
30
50 40 30
23.3
23
20
20
10
10
0
0
0
PCA3≤35<50
PCA3≥50
35≤PCA3<50
PCA3≥50
Gleason <7 <33% affected cylinders
≥33% affected cylinders
PCA3<50
PCA3<50
PCA3≥50
Total (n)
<33
12
26
38
≥33
0
13
13
Total (n)
12
39
51
PCA3≥50
Total
<7
10
10
20
≥7
3
33
36
Total
13
43
56
Gleason score
Percentage of affected cylinders
Gleason ≥7
(Fisher’s p=0.002)
Figure 2 Distribution of the Gleason score depending on the s-PCA3 in patients with diagnosis of PA (n = 56).
(Fisher’s p=0.023)
Figure 1 Percentage of affected cylinders depending on the s-PCA3 in patients with diagnosis of PA (n = 51).
significant relationship with age, the rate of serum PSA, the presence of prostatitis, or prostate volume. Of the total samples analyzed, 365 had a negative s-PCA3, corresponding to 328 patients. In these cases, accounting for 54.5% of the samples and 54% of the patients, no biopsy was indicated. Also, 293 samples (43.7%) showed a positive s-PCA3 and, in these cases, making a subsequent biopsy was recommended. Of the 124 biopsies studied, 56 (45.2%) were diagnosed with PA, 11 (8.8%) with ASAP, 10 (8.0%) with HGPIN, and 47 (38%) with mildness or normality. By placing the cut-off point of s-PCA3 at 100 or higher, the percentage of PA increased to 57.2%, while the percentages of ASAP and HGPIN did not change significantly (11.4 and 8.6%, respectively), and the percentage of biopsies with no evidence of malignancy decreased to 22.8%. The probability of positive biopsy increases with s-PCA3 (p = 0.040). The number of cylinders involved in the biopsy could be assessed in 51 cases; the remaining 5 were surgical resection samples. Comparing the percentage of cylinders affected in the biopsy (less than 33% vs. equal to or more than 33%) with the s-PCA3, it is shown that by increasing the s-PCA3, the percentage of affected cylinders increases and their relationship is statistically significant (p = 0.023) (Fig. 1). No patient with PA and s-PCA3 between 35 and 50 showed more than 33% of the studied cylinders involved. The s-PCA3 and Gleason score also showed a statistically significant relationship (p = 0.002). Only 23% of the patients with PA and s-PCA3 between 35 and 50 were high grade (Gleason ≥7). However, 76.7% of the patients with PA and s -PCA3 exceeding or equal to 50 had a Gleason ≥7 (Fig. 2). Selecting the best log-linear model including the s-PCA3 variables, Gleason score, and percentage of biopsy cylinders affected in the biopsy retained in the final equation the interaction between the s-PCA3 and Gleason score (p = 0.00006), confirming the relationship detected with the Fisher’s exact test. The model obtained also maintained the
s-PCA3 interaction and the percentage of affected cylinders (p = 0.004) (Fig. 3).
Discussion In our series, the s-PCA3 was independent of the PSA, patient age, prostate volume, and the presence of prostatitis, according to what was published in previous studies.21,22 The incorporation of the determination of s-PCA3 in PA screening significantly reduced the indication for biopsy (54%), and in patients with positive s-PCA3, the positivity of the biopsy markedly increased (45.2%) compared to that PCA3 vs gleason vs % affected cylinders 35≤Pca3≤49
50≤Pca3
% cyldr.≥33%
% cyldr.<33%
0
5
10
15
20
25
0
5
10
15
20
25
Number of cases Gleason<7 Gleason≥7
Figure 3 Bar chart with variables s-PCA3, Gleason score, and percentage of cylinders affected in the biopsy, represented simultaneously.
154 obtained only in accordance with the figures of the PSA. The s-PCA3 showed a statistically significant relationship with the Gleason score and the percentage of affected cylinders. These results agree with those reported in previously published meta-analyses and reviews.11,13 In the only previous study on Spanish population, the rates of diagnostic accuracy of PCA3 in clinical settings were assessed, showing as well that their incorporation improves screening and reduces the number of biopsies indicated. Using a cut-off point of 100 for the s-PCA3, we obtained, in our series, 57.2% of biopsies with PA, a slightly higher percentage than that previously reported in Spanish population.14 Additionally, we checked, in our population, the relationship between the PCA3 and the parameters associated with tumor aggressiveness such as the Gleason score and percentage of cylinders affected in the biopsy. In none of the biopsies in our series of patients with s-PCA between 35 and 50, involvement was observed in over 33% of the cylinders, and also, in this same interval of s-PCA3, 77% of the detected PAs had a Gleason below 7. These figures are reversed in patients with PA and s-PCA3 higher than or equal to 50 and are consistent with those reported by van Poppel et al. in a European study that enrolled more than 1,000 patients.16 These data indicate that in patients with a diagnosis of PA but with s-PCA3 lower than 50, the lower grade and stage PAs predominate, potential candidates for conservative clinical approach. The 54% reduction in the indication for biopsy obtained in our study is a first argument which, together with the data from previous studies,11,14 supports the introduction of the PCA3 as a complementary test in the management of patients suspected of PA. A second argument is provided by the increased efficiency of PCA3-guided biopsies.23,24 If with the serum PSA and the urological examination the percentage of positive biopsies is lower than 40%,25 with the protocol of indications of PCA3 of our study, using a cutoff point greater than or equal to 35, it rises to 45.2% and reaches 54% if we include the cases labeled as ASAP that, in our laboratory, correspond to proliferations with morphology and immunophenotype of PA but underrepresented in the biopsies to support a conclusive diagnosis. Additionally, the predictive value of grade and stage of s-PCA3 brings a new resource to assess the potential clinical significance of the PA before conducting the biopsy and selecting the suitable clinical behavior. Regardless of the importance of consolidating these results prospectively expanding the volume of cases studies and their clinical follow-up, the currently available information suggests that the s-PCA3 can provide new ways of solving the problem that the screening for prostate cancer is generating with the classic clinical and biochemical resources, of suboptimal specificity and efficiency. The use of serum PSA as an element of PA screening has not only enabled us to deepen in the knowledge of its biological behavior, but it has also shown that our limitations to predict it, in order to reduce mortality, entail, in a significant proportion of cases, diagnoses and treatments of doubtful efficiency. The incorporation of the PCA3 test and other biomarkers studied, together with the technological optimization of the radiological and surgical techniques, should provide, in the short term, new criteria for selecting the ideal clinical behavior in patients with suspected PA and the best therapeutic option in those already diagnosed by biopsy.
N. Rodón et al.
Conflict of interest The authors declare that they have no conflict of interest.
Acknowledgements To Eva Torija and Ana Guardiola for secretarial assistance. To Beatriz García-Peláez, Verónica Serrano, Patricia García, Bego Martínez, Nuria Arraiza and Cora Ferrer for technical assistance.
References 1. Schröder FH, Hugosson J, Roobol MJ, Tammela TL, Ciatto S, Nelen V, et al. Screening and prostate-cancer mortality in a randomized European study. N Engl J Med. 2009;360:1320---8. 2. Pérez-Lanzac-Lorca A, Barco-Sánchez A, Romero E, MartínezPeinado A, López-Elorza F, Sánchez-Sánchez E, et al. Correlación del cociente PSA complex/PSA total con el cociente PSA libre/PSA total, sensibilidad y especificidad de ambos marcadores para el diagnóstico del cáncer de próstata. Actas Urol Esp. 2013;37:498---503. 3. Baumann C, Behbahani TE. EAU guidelines for prostate cancer: to screen or not to screen? Actas Urol Esp. 2012;36:1. 4. Wilt TJ, Brawer MK, Jones KM, Barry MJ, Aronson WJ, Fox S, et al. Radical prostatectomy versus observation for localized prostate cancer. N Engl J Med. 2012;367:203---13. 5. Coley CM, Barry MJ, Fleming C, Mulley AG. Early detection of prostate cancer. Part I: prior probability and effectiveness of tests. The American College of Physicians. Ann Intern Med. 1997;126:394---406. 6. Andriole GL, Crawford ED, Grubb 3rd RL, Buys SS, Chia D, Church TR, et al. Mortality results from a randomized prostate-cancer screening trial. N Engl J Med. 2009;360:1310---9. 7. Moyer VA, U.S. Preventive Service Task Force. Screening for prostate cancer: U.S. Preventive Services Task Force recommendation statement. Ann Intern Med. 2012;157:120---34. 8. Fernández-Serra A, Rubio-Briones J, García-Casado Z, Solsona JA, López-Guerrero JA. Prostate cancer: the revolution of the fusion genes. Actas Urol Esp. 2011;35:420---8. 9. Ribeiro da Silva MN, Tobias-Machado M, Lima-Pompeo AC, Reis LO, da Silva Pinhal MA. Prostate cancer: promising biomarkers related to aggressive disease. Actas Urol Esp. 2012;36:484---90. 10. Bussemakers MJ, van Bokhoven A, Verhaegh GW, Smit FP, Karthaus HF, Schalken JA, et al. DD3: a new prostate-specific gene, highly overexpressed in prostate cancer. Cancer Res. 1999;59:5975---9. 11. Hessels D, Schalken JA. The use of PCA3 in the diagnosis of prostate cancer. Nat Rev Urol. 2009;6:255---61. 12. Groskopf J, Aubin SM, Deras IL, Blase A, Bodruq S, Clark C, et al. APTIMA PCA3 molecular urine test: development of a method to aid in the diagnosis of prostate cancer. Clin Chem. 2006;52:1089---95. 13. Ruiz-Aragón J, Márquez-Peláez S. Evaluación del test PCA3 para el diagnóstico de cáncer de próstata: revisión sistemática y metanálisis. Actas Urol Esp. 2010;34:346---55. 14. Rubio-Briones J, Fernández-Serra A, Ramírez M, Rubio L, Collado A, Casanova J, et al. Outcomes of expanded use of PCA3 testing in a Spanish population with clinical suspicion of prostate cancer. Actas Urol Esp. 2011;35:589---96. 15. Day JR, Jost M, Reynolds MA, Groskopf J, Rittenhouse H. PCA3: from basic molecular science to the clinical lab. Cancer Lett. 2011;301:1---6. 16. Van Poppel H, Haese A, Graefen M, de la Taille A, Irani J, de Reijke T, et al. The relationship between Prostate
Diagnostic and predictive value of PCA3 gene in patients with altered prostatic specific antigen
17.
18.
19.
20.
CAncer gene 3 (PCA3) and prostate cancer significance. BJU Int. 2012;109:360---6. Heidenreich A, Bellmunt J, Bolla M, Joniau S, Manson M, Matveev V, et al. Guía de la EAU sobre el cáncer de próstata. Parte I: cribado, diagnóstico y tratamiento del cáncer clínicamente localizado. Actas Urol Esp. 2011;35: 501---14. Epstein JI, Allsbrook Jr WC, Amin MB, Egevad LL, ISUP Grading Committe. The 2005 International Society of Urological Pathology (ISUP) Consensus Conference on Gleason Grading of Prostatic Carcinoma. Am J Surg Pathol. 2005;29: 1228---42. Friendly M. Working with categorical data with R and the vcd and vcdExtra packages. York University, Toronto [publicado 26 Jun 2013]. Disponible en: http://cran.r-project.org/ web/packages/vcdExtra/vignettes/vcd-tutorial.pdf R Development Core Team. R: a language and environment for statistical computing. Vienna, Austria: R Foundation for Statistical Computing; 2011. ISBN 3-900051-07-0.
155
21. De Luca S, Passera R, Milillo A, Coda R, Randone DF. Histological chronic prostatitis and high-grade prostate intra-epithelial neoplasia do not influence urinary prostate cancer gene 3 score. BJU Int. 2012;110:E778---82. 22. Haese A, de la Taille A, van Poppel H, Marberger M, Stenzl A, Mulders PF, et al. Clinical utility of the PCA3 urine assay in European men scheduled for repeat biopsy. Eur Urol. 2008;54:1081---8. 23. De la Taille A, Irani J, Graefen M, Chun F, de Reijke T, Kil P, et al. Clinical evaluation of the PCA3 assay in guiding initial biopsy decisions. J Urol. 2011;185:2119---25. 24. Shariat SF, Scherr DS, Gupta A, Bianco Jr FJ, Karakiewicz PI, Zeltser IS, et al. Emerging biomarkers for prostate cancer diagnosis, staging, and prognosis. Arch Esp Urol. 2011;64:681---94. 25. Remzi M, Fong YK, Dobrovits M, Anagnostou T, Seitz C, Waldert M, et al. The Vienna nomogram: validation of a novel biopsy strategy defining the optimal number of cores based on patient age and total prostate volume. J Urol. 2005;174:1256---60, discussion 1260-1251; author reply 1261.
Actas Urológicas Españolas www.elsevier.es/actasuro
ORIGINAL ARTICLE
Diagnostic and predictive value of urine PCA3 gene expression for the clinical management of patients with altered prostate-specific antigen夽 N. Rodón a,∗ , I. Trías a,b , M. Verdú a,b , R. Román a , A. Domínguez c , M. Calvo d , J.M. Banus c , A.M. Ballesta e , M.L. Maestro f , X. Puig a,b,g a
BIOPAT, Biopatologia Molecular, S.L., Grup Assistència, Barcelona, Spain Histopat Laboratoris, S.L., Barcelona, Spain c ICUN, Institut Català d’Urologia i Nefrologia, Barcelona, Spain d Departamento de Estadística, Facultad de Biología, Universidad de Barcelona, Barcelona, Spain e Analiza. S. Análisis Clínicos, Hospital Moncloa, Madrid, Spain f Unidad de Genómica y Reproducción Asistida, Hospital Clínico Universitario San Carlos, Madrid, Spain g Hospital de Barcelona, SCIAS, Grup Assistència, Barcelona, Spain b
Received 25 April 2013; accepted 17 July 2013 Available online 17 December 2013
KEYWORDS PCA3; Prostate-specific antigen; Screening; Prostate cancer; Gleason; Tumor aggressiveness
Abstract Objective: Analyze the impact of the introduction of the study of PCA3 gene in post-prostatic massage urine in the clinical management of patients with altered PSA, evaluating its diagnostic ability and predictive value of tumor aggressiveness. Methods: Observational, prospective, multicenter study of patients with suspected prostate cancer (PC) candidates for biopsy. We present a series of 670 consecutive samples of urine collected post-prostatic massage for three years in which we determined the «PCA3 score» (s-PCA3). Biopsy was only indicated in cases with s-positive PCA3. Results: The s-PCA3 was positive in 43.7% of samples. In the 124 biopsies performed, the incidence of PC or atypical small acinar proliferation was 54%, reaching 68.6% in s-PCA3 ≥ 100. Statistically significant relationship between the s-PCA3 and tumor grade was demonstrated. In cases with s-PCA3 between 35 and 50 only 23% of PC were high grade (Gleason ≥ 7), compared to 76.7% in cases with s-PCA3 over 50. There was a statistically significant correlation between s-PCA3 and cylinders affected. Both relationships were confirmed by applying a log-linear model. Conclusions: The incorporation of PCA3 can avoid the need for biopsies in 54% of patients. s-PCA3 positivity increases the likelihood of a positive biopsy, especially in higher s-PCA3 100 (68.6%). s-PCA3 is also an indicator of tumor aggressiveness and provides essential information in making treatment decisions. © 2013 AEU. Published by Elsevier España, S.L. All rights reserved.
夽 Please cite this article as: Rodón N, Trías I, Verdú M, Román R, Domínguez A, Calvo M, et al. Valor diagnóstico y predictivo del estudio en orina del gen PCA3 para el manejo clínico de pacientes con PSA alterado. Actas Urol Esp. 2013;38:150---155. ∗ Corresponding author. E-mail address: [email protected] (N. Rodón).
2173-5786/$ – see front matter © 2013 AEU. Published by Elsevier España, S.L. All rights reserved.
Diagnostic and predictive value of PCA3 gene in patients with altered prostatic specific antigen
PALABRAS CLAVE PCA3; Antígeno prostático específico; Cribado; Cáncer de próstata; Gleason; Agresividad tumoral
151
Valor diagnóstico y predictivo del estudio en orina del gen PCA3 para el manejo clínico de pacientes con PSA alterado Resumen Objetivo: Analizar el impacto de la introducción del estudio del gen PCA3 en orina posmasaje prostático, en el manejo clínico de pacientes con PSA alterado, valorando su capacidad diagnóstica y predictiva de agresividad tumoral. Métodos: Estudio observacional, prospectivo y multicéntrico de pacientes con sospecha de adenocarcinoma de próstata (AP) candidatos a biopsia. Presentamos una serie de 670 muestras consecutivas de orina posmasaje prostático recogidas durante 3 a˜ nos en las que se determinó el «score de PCA3» (s-PCA3). Se indicó biopsia únicamente en los casos con s-PCA3 positivo. Resultados: El s-PCA3 fue positivo en el 43,7% de las muestras. En las 124 biopsias realizadas, la incidencia de AP o proliferación microacinar atípica fue del 54%, llegando al 68,6% en sPCA3 ≥ 100. Demostramos relación estadísticamente significativa entre el s-PCA3 y el grado tumoral. En los casos con s-PCA3 positivo pero inferior a 50, solo el 23% de AP fueron de alto grado (Gleason ≥ 7); en cambio, en los casos con s-PCA3 superior a 50 lo fueron el 76,7%. La relación entre el s-PCA3 y el porcentaje de cilindros afectados en la biopsia fue significativa. Ambas relaciones se confirmaron al aplicar el modelo log-lineal. Conclusiones: La incorporación del PCA3 permite prescindir de la biopsia en un 54% de los pacientes. La positividad del s-PCA3 aumenta la probabilidad de biopsia positiva, especialmente en s-PCA3 superior a 100 (68,6%). El s-PCA3 es también un indicador de agresividad tumoral y aporta información esencial en la toma de decisiones terapéuticas. © 2013 AEU. Publicado por Elsevier España, S.L. Todos los derechos reservados.
Introduction The generalization of the analysis of prostate-specific antigen (PSA) in serum for early detection of prostate adenocarcinoma (PA) has generated a significant increase in the number of prostate biopsies; however, the low specificity of PSA generates a significant volume of indications for biopsy in patients with benign pathologies. About 75% of the biopsies are negative.1 In the group of patients with altered PSA values but not above 10 ng/ml, which is known as the ‘gray area’ of the diagnosis,2 patients with benign prostatic hyperplasia (BPH), prostatitis, and PA are mixed. In the first 2 cases, the biopsy will be negative and it will be repeated if the PSA levels persist. In the third case, the patients are diagnosed with PA but, in many cases, they are questionable clinical significance carcinomas and in elderly patients.3 In this context, radical surgical treatment does not necessarily modify the patient’s life expectancy.4 In autopsy series reviews, the incidental finding of hidden PA in 30% of patients from 50 to 70 and in 70% after 70 is reported.5 These neoplasms produce no symptoms and do not compromise the patient’s life, but they are able to elevate the serum PSA and trigger the clinical protocol of biopsy and surgical treatment. Currently, there is consensus on the problems generated by PA screening with PSA, and the problems it creates are considered more relevant than the benefits in terms of reducing mortality.1,6 The U.S. agency of preventive medicine (U.S. Preventive Services Task Force) recently issued a grade D recommendation which advises against PA screening with PSA,7 and it has raised reflections and discussions in subsequent articles, which reflect the need to rethink about the current protocols for screening, diagnosis, and treatment of PA and deepen in the search for new specific biomarkers predictive of PA.8,9
The PCA3 gene, widely studied in recent years, is probably one of the first biomarkers which is already under clinical application. This gene, located on chromosome 9q21-22, is transcribed only in prostate tissue and is strongly overexpressed in malignant prostate tissue compared to benign or normal adjacent one.10 In contrast to the PSA, overexpression of PCA3 is specific of PA and independent from prostate volume and the presence of BPH.11 Since 2006, a diagnostic kit of clinical application for the detection of mRNA levels of the PCA3 gene in urine samples collected after prostate massage has been marketed, with which a PCA3 score is obtained (s-PCA3).12 A meta-analysis published in 2010 attributes to PCA3 a negative predictive value of 87.8---98%, recognizing some diagnostic accuracy rates acceptable for use in the diagnosis of PA.13 The test, moreover, is easy to implement both in the clinical protocol and in the laboratory. From the first studies carried out, a directly proportional relationship between the s-PCA3 and the percentage of subsequent positive biopsies became clear.11,14 In some of the most recent studies, we also found a relationship between the s-PCA3 and different determinants of tumor aggressiveness such as the Gleason score, the tumor volume, and extracapsular invasion.11,15,16 The European Association of Urology, in its 2010 Guidelines, includes the use of biomarkers complementary to the PSA, among which the PCA3 stands out due to its higher specificity and sensitivity.17 In this paper we analyze the impact of the introduction of PCA3 in the clinical management of patients with altered PSA, with a special focus on reducing the indication for prostate biopsy and its alleged predictive ability of tumor aggressiveness in patients already diagnosed with PA.
152
Materials and methods We performed an observational, prospective, and multicenter study of patients with suspected PA candidates for prostate biopsy. A protocol of indications with informed consent was established and submitted to an Ethics Committee.
Samples and patients 670 urine samples were included in the study after prostate massage corresponding to 598 consecutive patients presenting to the urologist consultation between October 2009 and November 2012. 66 patients had different urine samples studied throughout their clinical follow-up. For the comparative statistical analyses with other clinical parameters (age, serum PSA rate, presence of prostatitis, and prostate volume) we only used the last study of PCA3, excluding previous ones. This resulted in the exclusion of 72 samples. All the patients met some of the following criteria: elevated PSA and prior negative biopsy, altered PSA but lower than 10 ng/ml without prior biopsy, altered PSA and well-known benign prostatic disease (prostatitis, BPH) and high PSA in patients with comorbidity associated with increased risk in biopsy practice. We excluded patients with previous diagnosis of PA. Age, the value of total PSA (ng/ml), and the prostate volume (cc) of the patients included were recorded. In the cases in which the biopsy was performed, the number of cylinders obtained and, when PA was diagnosed, the number of affected cylinders, the percentage of tumor, and Gleason score were recorded.
PCA3 determination The first voided urine after DRE with prostatic massage, consisting in 3 palpations per lobe, was collected, and the s-PCA3 was determined with the ProgensaTM PCA3Assay (GenProbe) kit. By means of quantification of PCA3 and PSA mRNA, an s-PCA3 was calculated according to the formula: (PCA3 mRNA/PSA mRNA) × 1000. s-PCA3 higher than or equal to 35 was considered positive, interpreted as a high probability of detecting PA in a subsequent biopsy.12 Biopsy was indicated only in cases with positive s-PCA3.
Prostate biopsy protocol Prostate biopsies were performed in the operating room under anesthesia with sedation, ultrasound-guided, and transrectally, obtaining a minimum of 5 cylinders per lobe, with additional samples in the case of finding suspicious nodules on the digital rectal examination or the ultrasound. In those patients who had already had prostate biopsies performed previously, these biopsies were performed by saturation, obtaining a minimum of 10 cylinders per lobe.
Histopathological study Two pathologists performed the histopathological study independently on serial sections (5 microns) of paraffinembedded tissue and stained with hematoxylin-eosin
N. Rodón et al. complemented with immunohistochemical study with racemase (AMACR) and basal cell markers (p63 and 34E12 cytokeratin) when it was necessary. The diagnostic entities covered were PA, atypical small acinar proliferation (ASAP), high-grade prostatic intraepithelial neoplasia (HGPIN), chronic prostatitis, other non-neoplastic processes, or normality. In the PA cases, the grade was determined following the Gleason patterns18 and the number and percentage of affected cylinders were recorded. The ASAP category was reserved for microglandular proliferations with morphological and immunohistochemical criteria of PA (loss of basal layer and/or expression of AMACR) but underrepresented in biopsies to support a definitive diagnosis of PA.
Statistical analysis The possible associations between pairs of categorical variables were analyzed using Fisher’s exact test, while for the continuous variables we used the simple linear regression model. Finally, to determine the association between a shortlist of three categorical variables, the best loglinear model was established using the stepwise algorithm and the Akaike information criterion. The algorithm was based on the full model, i.e., with all possible interactions among the 3 variables.19 All the statistical analyses were performed with the R v.2.15.1 package (©2012, RDevelopmentCoreTeam).20
Results The study included 670 urine samples of 598 patients with a mean age of 65.7 years (standard deviation of 8). In the majority of cases (90.6%), the PCA3 test was indicated due to elevated PSA and negative previous biopsy or due to alteration of PSA not exceeding 10 ng/ml (Table 1). The information rate of the s-PCA3 was 98.2% and the average 47.3. The medians and ranges of the numeric variables studied are shown in Table 2. The s-PCA3 showed no statistically Table 1
Indications for the study of PCA3.
Indication
Number of samples (%)
Negative prior biopsy Elevated PSA ≤ 10 ng/ml Risk factors for biopsy Known benign pathology Not specified
337 (50.3) 270 (40.3) 27 (4.0) 20 (3.0) 16 (2.4)
Total
670 (100)
Table 2
Description of the variables studied. Median (minimum-maximum)
Total PSA (ng/ml) Prostate volume (cc) Number of cylinders studied Percentage of affected cylinders (%)
6 (0.6---134) 53 (10---196) 11 (4---33) 20 (4---100)
Diagnostic and predictive value of PCA3 gene in patients with altered prostatic specific antigen 100
100
90
90
80
80
70 66.6
60 50 40
76.7
77
60 % of cases
70 % of cases
153
33.4
30
50 40 30
23.3
23
20
20
10
10
0
0
0
PCA3≤35<50
PCA3≥50
35≤PCA3<50
PCA3≥50
Gleason <7 <33% affected cylinders
≥33% affected cylinders
PCA3<50
PCA3<50
PCA3≥50
Total (n)
<33
12
26
38
≥33
0
13
13
Total (n)
12
39
51
PCA3≥50
Total
<7
10
10
20
≥7
3
33
36
Total
13
43
56
Gleason score
Percentage of affected cylinders
Gleason ≥7
(Fisher’s p=0.002)
Figure 2 Distribution of the Gleason score depending on the s-PCA3 in patients with diagnosis of PA (n = 56).
(Fisher’s p=0.023)
Figure 1 Percentage of affected cylinders depending on the s-PCA3 in patients with diagnosis of PA (n = 51).
significant relationship with age, the rate of serum PSA, the presence of prostatitis, or prostate volume. Of the total samples analyzed, 365 had a negative s-PCA3, corresponding to 328 patients. In these cases, accounting for 54.5% of the samples and 54% of the patients, no biopsy was indicated. Also, 293 samples (43.7%) showed a positive s-PCA3 and, in these cases, making a subsequent biopsy was recommended. Of the 124 biopsies studied, 56 (45.2%) were diagnosed with PA, 11 (8.8%) with ASAP, 10 (8.0%) with HGPIN, and 47 (38%) with mildness or normality. By placing the cut-off point of s-PCA3 at 100 or higher, the percentage of PA increased to 57.2%, while the percentages of ASAP and HGPIN did not change significantly (11.4 and 8.6%, respectively), and the percentage of biopsies with no evidence of malignancy decreased to 22.8%. The probability of positive biopsy increases with s-PCA3 (p = 0.040). The number of cylinders involved in the biopsy could be assessed in 51 cases; the remaining 5 were surgical resection samples. Comparing the percentage of cylinders affected in the biopsy (less than 33% vs. equal to or more than 33%) with the s-PCA3, it is shown that by increasing the s-PCA3, the percentage of affected cylinders increases and their relationship is statistically significant (p = 0.023) (Fig. 1). No patient with PA and s-PCA3 between 35 and 50 showed more than 33% of the studied cylinders involved. The s-PCA3 and Gleason score also showed a statistically significant relationship (p = 0.002). Only 23% of the patients with PA and s-PCA3 between 35 and 50 were high grade (Gleason ≥7). However, 76.7% of the patients with PA and s -PCA3 exceeding or equal to 50 had a Gleason ≥7 (Fig. 2). Selecting the best log-linear model including the s-PCA3 variables, Gleason score, and percentage of biopsy cylinders affected in the biopsy retained in the final equation the interaction between the s-PCA3 and Gleason score (p = 0.00006), confirming the relationship detected with the Fisher’s exact test. The model obtained also maintained the
s-PCA3 interaction and the percentage of affected cylinders (p = 0.004) (Fig. 3).
Discussion In our series, the s-PCA3 was independent of the PSA, patient age, prostate volume, and the presence of prostatitis, according to what was published in previous studies.21,22 The incorporation of the determination of s-PCA3 in PA screening significantly reduced the indication for biopsy (54%), and in patients with positive s-PCA3, the positivity of the biopsy markedly increased (45.2%) compared to that PCA3 vs gleason vs % affected cylinders 35≤Pca3≤49
50≤Pca3
% cyldr.≥33%
% cyldr.<33%
0
5
10
15
20
25
0
5
10
15
20
25
Number of cases Gleason<7 Gleason≥7
Figure 3 Bar chart with variables s-PCA3, Gleason score, and percentage of cylinders affected in the biopsy, represented simultaneously.
154 obtained only in accordance with the figures of the PSA. The s-PCA3 showed a statistically significant relationship with the Gleason score and the percentage of affected cylinders. These results agree with those reported in previously published meta-analyses and reviews.11,13 In the only previous study on Spanish population, the rates of diagnostic accuracy of PCA3 in clinical settings were assessed, showing as well that their incorporation improves screening and reduces the number of biopsies indicated. Using a cut-off point of 100 for the s-PCA3, we obtained, in our series, 57.2% of biopsies with PA, a slightly higher percentage than that previously reported in Spanish population.14 Additionally, we checked, in our population, the relationship between the PCA3 and the parameters associated with tumor aggressiveness such as the Gleason score and percentage of cylinders affected in the biopsy. In none of the biopsies in our series of patients with s-PCA between 35 and 50, involvement was observed in over 33% of the cylinders, and also, in this same interval of s-PCA3, 77% of the detected PAs had a Gleason below 7. These figures are reversed in patients with PA and s-PCA3 higher than or equal to 50 and are consistent with those reported by van Poppel et al. in a European study that enrolled more than 1,000 patients.16 These data indicate that in patients with a diagnosis of PA but with s-PCA3 lower than 50, the lower grade and stage PAs predominate, potential candidates for conservative clinical approach. The 54% reduction in the indication for biopsy obtained in our study is a first argument which, together with the data from previous studies,11,14 supports the introduction of the PCA3 as a complementary test in the management of patients suspected of PA. A second argument is provided by the increased efficiency of PCA3-guided biopsies.23,24 If with the serum PSA and the urological examination the percentage of positive biopsies is lower than 40%,25 with the protocol of indications of PCA3 of our study, using a cutoff point greater than or equal to 35, it rises to 45.2% and reaches 54% if we include the cases labeled as ASAP that, in our laboratory, correspond to proliferations with morphology and immunophenotype of PA but underrepresented in the biopsies to support a conclusive diagnosis. Additionally, the predictive value of grade and stage of s-PCA3 brings a new resource to assess the potential clinical significance of the PA before conducting the biopsy and selecting the suitable clinical behavior. Regardless of the importance of consolidating these results prospectively expanding the volume of cases studies and their clinical follow-up, the currently available information suggests that the s-PCA3 can provide new ways of solving the problem that the screening for prostate cancer is generating with the classic clinical and biochemical resources, of suboptimal specificity and efficiency. The use of serum PSA as an element of PA screening has not only enabled us to deepen in the knowledge of its biological behavior, but it has also shown that our limitations to predict it, in order to reduce mortality, entail, in a significant proportion of cases, diagnoses and treatments of doubtful efficiency. The incorporation of the PCA3 test and other biomarkers studied, together with the technological optimization of the radiological and surgical techniques, should provide, in the short term, new criteria for selecting the ideal clinical behavior in patients with suspected PA and the best therapeutic option in those already diagnosed by biopsy.
N. Rodón et al.
Conflict of interest The authors declare that they have no conflict of interest.
Acknowledgements To Eva Torija and Ana Guardiola for secretarial assistance. To Beatriz García-Peláez, Verónica Serrano, Patricia García, Bego Martínez, Nuria Arraiza and Cora Ferrer for technical assistance.
References 1. Schröder FH, Hugosson J, Roobol MJ, Tammela TL, Ciatto S, Nelen V, et al. Screening and prostate-cancer mortality in a randomized European study. N Engl J Med. 2009;360:1320---8. 2. Pérez-Lanzac-Lorca A, Barco-Sánchez A, Romero E, MartínezPeinado A, López-Elorza F, Sánchez-Sánchez E, et al. Correlación del cociente PSA complex/PSA total con el cociente PSA libre/PSA total, sensibilidad y especificidad de ambos marcadores para el diagnóstico del cáncer de próstata. Actas Urol Esp. 2013;37:498---503. 3. Baumann C, Behbahani TE. EAU guidelines for prostate cancer: to screen or not to screen? Actas Urol Esp. 2012;36:1. 4. Wilt TJ, Brawer MK, Jones KM, Barry MJ, Aronson WJ, Fox S, et al. Radical prostatectomy versus observation for localized prostate cancer. N Engl J Med. 2012;367:203---13. 5. Coley CM, Barry MJ, Fleming C, Mulley AG. Early detection of prostate cancer. Part I: prior probability and effectiveness of tests. The American College of Physicians. Ann Intern Med. 1997;126:394---406. 6. Andriole GL, Crawford ED, Grubb 3rd RL, Buys SS, Chia D, Church TR, et al. Mortality results from a randomized prostate-cancer screening trial. N Engl J Med. 2009;360:1310---9. 7. Moyer VA, U.S. Preventive Service Task Force. Screening for prostate cancer: U.S. Preventive Services Task Force recommendation statement. Ann Intern Med. 2012;157:120---34. 8. Fernández-Serra A, Rubio-Briones J, García-Casado Z, Solsona JA, López-Guerrero JA. Prostate cancer: the revolution of the fusion genes. Actas Urol Esp. 2011;35:420---8. 9. Ribeiro da Silva MN, Tobias-Machado M, Lima-Pompeo AC, Reis LO, da Silva Pinhal MA. Prostate cancer: promising biomarkers related to aggressive disease. Actas Urol Esp. 2012;36:484---90. 10. Bussemakers MJ, van Bokhoven A, Verhaegh GW, Smit FP, Karthaus HF, Schalken JA, et al. DD3: a new prostate-specific gene, highly overexpressed in prostate cancer. Cancer Res. 1999;59:5975---9. 11. Hessels D, Schalken JA. The use of PCA3 in the diagnosis of prostate cancer. Nat Rev Urol. 2009;6:255---61. 12. Groskopf J, Aubin SM, Deras IL, Blase A, Bodruq S, Clark C, et al. APTIMA PCA3 molecular urine test: development of a method to aid in the diagnosis of prostate cancer. Clin Chem. 2006;52:1089---95. 13. Ruiz-Aragón J, Márquez-Peláez S. Evaluación del test PCA3 para el diagnóstico de cáncer de próstata: revisión sistemática y metanálisis. Actas Urol Esp. 2010;34:346---55. 14. Rubio-Briones J, Fernández-Serra A, Ramírez M, Rubio L, Collado A, Casanova J, et al. Outcomes of expanded use of PCA3 testing in a Spanish population with clinical suspicion of prostate cancer. Actas Urol Esp. 2011;35:589---96. 15. Day JR, Jost M, Reynolds MA, Groskopf J, Rittenhouse H. PCA3: from basic molecular science to the clinical lab. Cancer Lett. 2011;301:1---6. 16. Van Poppel H, Haese A, Graefen M, de la Taille A, Irani J, de Reijke T, et al. The relationship between Prostate
Diagnostic and predictive value of PCA3 gene in patients with altered prostatic specific antigen
17.
18.
19.
20.
CAncer gene 3 (PCA3) and prostate cancer significance. BJU Int. 2012;109:360---6. Heidenreich A, Bellmunt J, Bolla M, Joniau S, Manson M, Matveev V, et al. Guía de la EAU sobre el cáncer de próstata. Parte I: cribado, diagnóstico y tratamiento del cáncer clínicamente localizado. Actas Urol Esp. 2011;35: 501---14. Epstein JI, Allsbrook Jr WC, Amin MB, Egevad LL, ISUP Grading Committe. The 2005 International Society of Urological Pathology (ISUP) Consensus Conference on Gleason Grading of Prostatic Carcinoma. Am J Surg Pathol. 2005;29: 1228---42. Friendly M. Working with categorical data with R and the vcd and vcdExtra packages. York University, Toronto [publicado 26 Jun 2013]. Disponible en: http://cran.r-project.org/ web/packages/vcdExtra/vignettes/vcd-tutorial.pdf R Development Core Team. R: a language and environment for statistical computing. Vienna, Austria: R Foundation for Statistical Computing; 2011. ISBN 3-900051-07-0.
155
21. De Luca S, Passera R, Milillo A, Coda R, Randone DF. Histological chronic prostatitis and high-grade prostate intra-epithelial neoplasia do not influence urinary prostate cancer gene 3 score. BJU Int. 2012;110:E778---82. 22. Haese A, de la Taille A, van Poppel H, Marberger M, Stenzl A, Mulders PF, et al. Clinical utility of the PCA3 urine assay in European men scheduled for repeat biopsy. Eur Urol. 2008;54:1081---8. 23. De la Taille A, Irani J, Graefen M, Chun F, de Reijke T, Kil P, et al. Clinical evaluation of the PCA3 assay in guiding initial biopsy decisions. J Urol. 2011;185:2119---25. 24. Shariat SF, Scherr DS, Gupta A, Bianco Jr FJ, Karakiewicz PI, Zeltser IS, et al. Emerging biomarkers for prostate cancer diagnosis, staging, and prognosis. Arch Esp Urol. 2011;64:681---94. 25. Remzi M, Fong YK, Dobrovits M, Anagnostou T, Seitz C, Waldert M, et al. The Vienna nomogram: validation of a novel biopsy strategy defining the optimal number of cores based on patient age and total prostate volume. J Urol. 2005;174:1256---60, discussion 1260-1251; author reply 1261.